State Identification of Power System Based on Visual Geometric Group Network

Abstract Accurate and timely identification of power system operation states are requisite for taking control strategies ensuring security. With the high grid penetration renewable energy generation wide application electronic equipment, is becoming more complicated facing uncertainties. Traditional model based state methods faced with challenges data driven have great prospects. This paper proposes a new data-driven method identifying operating on visual geometric group network VGG16 (16-layer Visual Geometry Group (VGG16). Firstly, Lasso algorithm used to select features realize dimensionality reduction. Then, weights loss functions adjusted according difficulty states. Finally, IEEE39 node as an example, effectiveness robustness proposed verified.